There is continued interest in development of antisense reagents (ASRs), including especially antisense oligonucleotides and small interfering RNAs, for experimental as well as therapeutic purposes. Optimization of ASRs begins with target site selection. Here we review protocols which have been developed to empirically determine effective target sites in RNAs. Such library selection technologies have demonstrated clear utility, and in vitro identification of sites has generally proven effective for cellular applications. A few groups are developing large combinatorial libraries and approaches to adapt use of such libraries to individual target RNAs, as well as learning algorithms to help with the optimization of target sites, particularly with respect to small interfering RNAs.